Chemistry of Winemaking - ACS Publications

well-adapted to a multi-scale operation and automation (27, 28). GLC is used ... acidity be made during processing and aging before marketing. The cur...
0 downloads 0 Views 2MB Size
6 The Present Status of Methods for Wine Analysis and Possible Future Trends

Downloaded by UNIV OF GUELPH LIBRARY on July 15, 2012 | http://pubs.acs.org Publication Date: November 1, 1974 | doi: 10.1021/ba-1974-0137.ch006

M A Y N A R D A. A M E R I N E Department of Viticulture and Enology, University of California, Davis, Calif. 90241

The analytical methods used in the wine industry (mainly of the United States) are summarized, and procedures required by legal and regulatory agencies are given in some detail. Ethanol, volatile acidity, and sulfur dioxide are the important constituents to detect. Methods for detecting and measuring metal, pesticide, and fungicide contaminants are also outlined. Other procedures required for rational winery operation include determination of acids, sugars, acetaldehyde, potassium, color pigments, tannins, etc. Finally, research in enology may require methods to determine compounds such as diacetyl, succinic acid, various esters, etc. Gas-liquid chromatography is being increasingly used in wine and must analyses.

^ m o n g Pasteur's many contributions to the wine industry were several simple methods of wine analysis, and his research has had a per­ manent influence on the wine industry. As a well-trained chemist he was certain that chemical analysis would reveal the nature of the process of alcoholic fermentation and the type and degree of spoilage related to it. In both cases he was correct. Even prior to Pasteur, alcohol content determination was important as a basis for local, import, and export taxes. Other important applica­ tions of accurate wine analysis have been to detect and to accurately determine food additives; now there are legal reasons for analyzing wines for sulfur dioxide, organic chloride or bromide, sodium, cyanide, diglucoside pigments, various insecticides, fungicides, etc. Winery con­ trol calls for analytical determination of iron, copper, protein, total acidity, p H , tartaric, malic and lactic acids, etc. Finally, quality control 134

In Chemistry of Winemaking; Webb, A.; Advances in Chemistry; American Chemical Society: Washington, DC, 1974.

Downloaded by UNIV OF GUELPH LIBRARY on July 15, 2012 | http://pubs.acs.org Publication Date: November 1, 1974 | doi: 10.1021/ba-1974-0137.ch006

6.

AMERINE

135

Wine Analysis

demands accurate chemical analysis to ensure uniformity within a brand or type for color, alcohol, total acidity, reducing sugar, etc. B y 1900 many analytical procedures were available. DujardinSalleron (France) not only codified these procedures but also produced the necessary equipment for them (1). Official methods of wine analysis were soon developed in France and many other countries, and our own Association of Official Agricultural Chemists began developing "tentative and official" methods of wine analysis as early as 1916 (2); these continue to the present (3). Official methods of analysis, both for reference and routine purposes, are given on the international field by the Office Inter­ national de la Vigne et du V i n (4) and by Amerine and Ough (5). For current American practices see Refs. 3, 5, 6, 7, 8, 9, 10, 11, 12, 12a. For European procedures see Refs. 4, 9,13,14,15,16,17,18,19, 20, 21, 22, 23. In many cases simple and only relatively accurate procedures are adequate for winery control purposes. Recently a number of 5-minute methods were developed in Europe (23a, 24,25), and Schmitt (26) notes that in these semimicro procedures (1 ml) care must be used to avoid loss of volatile material, such as alcohol, and to make accurate volumetric measurements with wines of varying viscosity. Losses during distillation must also be minimized. A small winery making a few determinations per year may prefer a simple but time-consuming procedure, while a large winery (100+ determinations/day) may save money without loss of accuracy by using a rapid procedure with expensive equipment. In some cases, particularly in Europe, governments require a particular method even if its not the most modern. Methods Required by Legal or Regulatory

Standards

International limits for contaminants (4) are: arsenic 0.2 mg/liter, volatile acidity 20 meq/liter for 10 vol % ethanol and 1 meq more for each per cent alcohol above 10%, lead 0.6 mg/liter, boron 80 mg/liter (as boric acid), bromine (total) 1 mg/liter (may be higher for wines from grapes of certain areas), bromine (organic) 0.0, fluorine 5 mg/liter, malvidin diglucoside 15 mg/liter, sodium 60 mg/liter (may be higher for wines from grapes of certain areas), and sulfate 1.5 grams/liter (as potassium sulfate). Ethanol. The accurate, reasonably rapid, and relatively inexpensive ethanol determination is the most important analytical procedure for the winery analyst. Not only is it required for legal reasons but also for winery control and research investigations. In spite of their inefficiency (and even inaccuracy), boiling point (ebulliometric) procedures still are used commonly by small wineries

In Chemistry of Winemaking; Webb, A.; Advances in Chemistry; American Chemical Society: Washington, DC, 1974.

136

C H E M I S T R Y OF W I N E M A K I N G

and by home wine makers. Specific gravity methods based on hydrometry are more popular and more accurate although they are time consuming. In both cases the simple equipment makes them ideal for operators who make relatively few determinations.

Downloaded by UNIV OF GUELPH LIBRARY on July 15, 2012 | http://pubs.acs.org Publication Date: November 1, 1974 | doi: 10.1021/ba-1974-0137.ch006

For large-scale operations dichromate oxidation procedures provide precision. Microdistillation and careful temperature and acidity control make dichromate oxidation procedures rapid, accurate, time saving, and well-adapted to a multi-scale operation and automation (27, 28). G L C is used increasingly as a rapid procedure for multi-determina­ tion laboratories. While the equipment is relatively expensive, careful operators get good results rapidly. N o doubt it w i l l be used more i n the future because many other determinations can be performed nearly simultaneously, particularly in research work; L i e et al. (29) used a G L C procedure requiring only three minutes per sample. The pycnometric procedure for determining ethanol is still the refer­ ence method for many countries although few laboratories use it routinely. In cases involving commercial transactions it is preferred, especially since the calculations can be simplified. Ethanol also may be determined using alcohol dehydrogenase and measuring the change in nicotinamide adenine dinucleotide ( N A D ) to the reduced form ( N A D H ) at 340 nm. +

alcohol dehydrogenase C H C H O H + NAD+ , C H C H O + N A D H + H+ 3

2

3

The solution is alkaline buffered, and semicarbazide is added to remove the acetaldehyde, forcing the reaction to the right. This method is par­ ticularly useful at low alcohol concentrations (30). Methanol. In grape wines and brandies methanol determination is rarely necessary except where excessive amounts of pomace are used; Italian regulations limit the methanol content for this reason. W i t h fruit wines and fruit brandies the legal maximum can easily be exceeded, and therefore an accurate determination of methanol is required. Methanol may be oxidized to formaldehyde, and the color developed with formaldehyde and roseaniline or chromotropic acid as indicator is used for methanol determination. G L C has also been used (31), and Martin et al. (32) found good agreement between the two procedures at higher methanol levels. Volatile A c i d i t y . Acetic acid is the primary acid formed during wine spoilage. Legal limits for it exist in all wine-producing countries, varying from 0.10 to 0.25% exclusive of sulfur dioxide and sorbic acid. The United States and California State limits are among the lowest. Good

In Chemistry of Winemaking; Webb, A.; Advances in Chemistry; American Chemical Society: Washington, DC, 1974.

Downloaded by UNIV OF GUELPH LIBRARY on July 15, 2012 | http://pubs.acs.org Publication Date: November 1, 1974 | doi: 10.1021/ba-1974-0137.ch006

6.

AMERINE

Wine Analysis

137

winery practice demands that accurate determinations of the volatile acidity be made during processing and aging before marketing. The current method (3, 4, 6, 22) involves steam distillation to sepa­ rate the volatile (primarily acetic) acids from the non-volatile (fixed) acids. Special equipment has been devised for this separation (6). Sulfurous and sorbic acid content can be corrected, or the sulfurous acid may be removed (33). Carbon dioxide must be removed so that it does not interfere with the test (6, 33). A n automated procedure is also available (34) which measures the volatile acids in the distillate at 450 nm using bromophenol blue. It is probable that G L C determination of acetic acid, as distinguished from other volatile acids (propionic, lactic, etc.), w i l l be used more often. A specific enzymatic procedure for acetate (35) reveals that only about two-thirds of the volatile acidity is acetic acid. However, the volatile acidity of commercial wines in modern wineries is usually well below the legal limits. S u l f u r Dioxide. The legal limits for total sulfur dioxide in wines varies from 200 to 350 mg/liter. In addition, the limit for sulfur dioxide not bound to aldehydes, polyphenolic compounds, etc. may be from 30 to 100 mg/liter. Winery control requires that the amount of sulfur dioxide present during processing and aging be carefully controlled, and increas­ ing concerns for public health reinforce this. For total sulfur dioxide determination, distillation procedures are preferred (3, 4, 6,22). Careful attention to details is necessary since the methods are empirical (16, 36, 37). Automated analysis (38) also can be used (up to 20 samples per h r ) . For a summary of investigations on sulfur dioxide determination see Wucherpfennig (39). The problem of accurately determining the non-bound (free) sulfur dioxide has not been satisfactorily solved (6, 7, 40). The best approach is to distill the sample in the absence of air and to recognize that the usual procedure overestimates the non-bound sulfur dioxide (41, 42, 43). For this reason some countries (—e.g., United States) do not specify a limit for non-bound sulfur dioxide. Nevertheless, in winery practice a limit is needed, and in countries where a legal limit exists, the determina­ tion is required. The direct Ripper iodometric titration is still used, but it is subject to error. In its place, direct iodate-iodide titration is used (44). This is followed by fixing the sulfur dioxide with glyoxal in a second sample and retitrating. The difference represents the free sulfur dioxide. The second titration roughly represents the amount of reduction and the amount of ascorbic acid present. Formulas for calculating the amount of sulfur dioxide to add in order to produce a predetermined level of free sulfur dioxide have been given by Stanescu (45).

In Chemistry of Winemaking; Webb, A.; Advances in Chemistry; American Chemical Society: Washington, DC, 1974.

Downloaded by UNIV OF GUELPH LIBRARY on July 15, 2012 | http://pubs.acs.org Publication Date: November 1, 1974 | doi: 10.1021/ba-1974-0137.ch006

138

C H E M I S T R Y OF W I N E M A K I N G

Metal Contaminants. There is increasing interest in the lead, sodium, and other metallic constituents of wine. The present lead limit is 0.6 mg/liter. W h i l e sodium is a normal constituent of wines, cation exchange resins may unduly increase its level. Suggested limits for sodium vary from 60 to 300 mg/liter in different countries. Wine as a low-sodium high-potassium beverage in diets for hypertension patients makes it ap­ parent that stricter limits on sodium may be applied. The suggested international limit for arsenic of 0.2 mg/liter should not be difficult to meet. Arsenate insecticides are not used i n the United States and are seldom used abroad. W h i l e simple tests for the presence of arsenic are available, its quantitative determination is time consuming, involving ashing, special equipment, and meticulous technique and cleanliness. Lead likewise should seldom if ever be found in wines approaching the suggested international limit of 0.6 mg/liter. The procedure by ashing and color development with dithiozone (or other reagent) is also time consuming and meticulous. Even with atomic absorption spectropho­ tometry there are losses in ashing and laboratory lead pick-up. Probably the best new attachments are those where the wine can be ashed directly in the apparatus. Boron (as boric acid) has a suggested international limit of 80 mg/liter. The probable source is soil or irrigation water, and several procedures are available for its determination. Fluorine (or fluorides) has a suggested international limit of 5 mg/liter. Fluoride pickup occurs only in the rare case where concrete tanks may have been treated with fluosilicate. W i t h the disappearance of concrete tanks and prohibition of fluosilicate which may contaminate food, this determination should be superfluous. It is done on an ashed sample distillation of fluosilicic acid with super-heated steam and titration with thorium nitrate in the presence of sodium alizarinsulfonate. Bromine (as bromide) may occasionally exceed the suggested 1 mg/liter limit in areas where grapes are grown on brackish soils. Here a higher limit can be applied since bromide is not toxic at this level. The determination involves ashing. Addition of Chloramine Τ in the presence of phenolsulfonphthalein leads to the formation of tetrabromophenolsulfonphthalein which is colored. Organic bromine has a zero tolerance because of the reported toxicity of monobromacetic acid. It can be extracted quantitatively with ether at p H 1. The determination is then the same as with total bromide (4). Sulfate (as potassium sulfate) has long been limited to reduce addi­ tion of calcium sulfate (plastering). The practice lowers the p H and is limited to warm climatic regions where the acidity is very low ( p H

In Chemistry of Winemaking; Webb, A.; Advances in Chemistry; American Chemical Society: Washington, DC, 1974.

Downloaded by UNIV OF GUELPH LIBRARY on July 15, 2012 | http://pubs.acs.org Publication Date: November 1, 1974 | doi: 10.1021/ba-1974-0137.ch006

6.

AMERINE

Wine Analysis

139

high). The suggested limit of 1.5 grams/liter is somewhat lower than the 2 and 3 grams/liter limit of various countries. The precipitation as barium sulfate requires boiling in the absence of air to remove sulfur dioxide. (The residue from the distillation pro­ cedure for sulfur dioxide may be used.) Some skill in handling the pre­ cipitate is required for accurate results. Complexometric titration after precipitation as lead sulfate (4) is possible. By using three solutions containing different amounts of barium chloride, it is possible to rapidly determine from the clouding whether a wine contains more or less than 0.7, 1.0, or 2.0 grams/liter of potassium sulfate. Nitrate is of interest as a possible indication of watering. M a n y potable water supplies contain 30 mg/liter of nitrate whereas normal wines contain about 4-5 mg/liter on the average (46,47,48). The present recommended procedure (49) is reduction of nitrate to nitrite followed by colorimetric determination with sulfanilic acid and a-naphthylamine. Other metals are rarely present in amounts approaching public health limits (see below for their determination in normal winery practice). Atomic absorption spectrophotometry is the method of preference (50, 51). For recent reviews of this procedure as applied to wines, see Refs. 52 and 53. X-ray fluorescence spectrophotometry has also been used (54). Diglucoside Anthocyan Pigments. N o limits on the amounts of d i ­ glucoside anthocyan pigments present in wines exist in the United States. In Europe a limit of 4 mg/liter of malvin has been suggested (22). The limit has more economic significance than any public health hazard. Paper and thin-layer chromatographic procedures are available when this determination is required (primarily in Germany) (4, 22, 55, 56, 57,58). Carbon Dioxide. The U.S. limit for carbon dioxide in non-sparkling wine is 2.77 grams/liter. Above this value the tax rises from 170 per gallon to $2.40 or $3.40. Other countries have less stringent limits. Obvi­ ously an accurate method is required, and several are available (4, 5, 6). A t present a simple enzymatic reaction using carbonic anhydrase is pre­ ferred. The bicarbonate ion is titrated with standard acid between p H 8.6 and 4.0. Carbonic anhydrase ensures that the carbonic acid is all in the bicarbonate form. A non-fading endpoint is thus obtained. Sugar and E x t r a c t . Glucose and fructose are the main reducing sugars present in musts and wines. Even when sucrose is present, it is hydrolyzed in a few days at the relatively low p H of the wine or by sucrase. In addition to legal restrictions on reducing sugars and sucrose in certain but not all countries, good winery practice requires accurate analyses in wines and proximate analyses in musts.

In Chemistry of Winemaking; Webb, A.; Advances in Chemistry; American Chemical Society: Washington, DC, 1974.

140

C H E M I S T R Y OF W I N E M A K I N G

The classical gravimetric copper procedures have been used for many years and are still the standard. Though they are accurate, they are time consuming. American enologists generally prefer the titrimetric Lane and Eynon (3,4,6) procedure. The elegant enzymatic procedure of Drawert and Kupper (59) per­ mits determination of glucose, fructose, and (when necessary) sucrose. Glucose reacts with adenosintriphosphate ( A T P ) in the presence of hexokinase ( H K ) to produce glucose-6-phosphate.

Downloaded by UNIV OF GUELPH LIBRARY on July 15, 2012 | http://pubs.acs.org Publication Date: November 1, 1974 | doi: 10.1021/ba-1974-0137.ch006

HK glucose + A T P ,

fructose-6-phosphate

A d d i n g glucose-6-phosphate dehydrogenase ( G - 6 - P D H ) in the presence of oxidized nicotinamide adenine trinucleotide ( N A D H ) yields gluconic acid-6-phosphate and an equivalent amount of the reduced form, N A D P H . The amount of N A D P H can be measured from its absorbance at 340 nm. For fructose the first reaction is: HK fructose + A T P .

fructose-6-phosphate

Fructose-6-phosphate is converted to glucose-6-phosphate in the presence of phosphoglucoseisomerase ( P G I ) . Trimethylsilyl derivatives of sugars can be made and separated by G L C (60). For musts, where 90% or more of the soluble solids are glucose and fructose, proximate procedures are usually sufficient—hydrometry or refractometry. Since only the sugar content up to 1 % is often needed, sugar test pills are useful (5, 6, 61, 62, 63). Rapid spectrophotometry procedures are sometimes convenient, especially if they do not require prior filtration or decoloration (62). Fraudulent addition of sucrose is very difficult to detect (22) because some sucrose is normally present, and added sucrose is rapidly hydrolyzed. Since commercial sucrose contains small amounts of unfermentable im­ purities, it has been suggested (60) that G L C would detect these i n wines. Synthetic sweetening agents can be detected by thin-layer chromatogra­ phy (64,65). The non-sugar soluble solids remaining in a table wine after dealcoholization are known as the extract. L o w sugar musts or watered and sugared musts have a low extract content. The minimum extract in the United States is 1.6 and 1.8 grams/100 m l for white and red table wines respectively. These limits are so low that wines from moderately watered musts w i l l meet them. While the concept of extract is relatively unambiguous, its precise definition is difficult. The problem is how to dealcoholize the wine with-

In Chemistry of Winemaking; Webb, A.; Advances in Chemistry; American Chemical Society: Washington, DC, 1974.

6.

AMERINE

Wine

141

Analysis

out loss of non-sugar soluble solids (lactic acid, glycerol, etc.). The empirical definition (4) is to specify a temperature ( 7 0 ° C ) , degree of vacuum (20-25 torr), period of time (1 hr plus cooling over sulfuric acid), and other conditions. The Taberié formula is usually used: d = d — d + 1.000 where d is the density of the residual, d is that of the wine (less the volatile acidity and sulfur dioxide), and d* that of the alcohol (both at 2 0 ° C ) . Tables for converting d to grams of extract per 100 m l are available, but the proper table to use is the Plato sucrose table, not the Ackermann empirical table (66). Pesticides and Fungicides. Modern pure food regulations require that the food processor be responsible for their finished products. Since so many pesticides and fungicides are used in agriculture, their detection and quantitative analysis are difficult (5, 22). Organophosphorus and chlorinated hydrocarbons are the most common pesticides. When G L C is used for halogens, electron capture or microcoulometric detectors are used; for phosphorus, a thermionic flame photometric detector is required. When the specific additive is known, simpler procedures can be devised (67). Some insecticides can be detected from their effect on cholinesterase activity (68). Prohibited Additives. Diethylpyrocarbonate ( D E P C ) is now pro­ hibited in the United States and other countries. Since a small but regular amount of diethyl carbonate is a constant byproduct of its use and is easily detectable with G L C , wines to which D E P C has been added are easily detected (69, 70, 71). Potassium ferrocyanide has long been used to remove excess copper and iron from wines. When not used in excess it appears effective and harmless, but if any ferrocyanide residue remains, cyanide may form. While the amounts produced by a slight excess would pose little danger, the blue precipitate and distinctive odor would be undesirable. Special equipment has been devised to detect free cyanide and ferrocyanide (4, 5, 6) as Prussian blue. The suggested limit is 1 mg/liter as cyanide. Hoppe and Romminger (72) devised a rapid procedure for free and bound cyanide, and Bates (73) gives a qualitative screening method sensitive to 0.05 mg/liter. Rarely are added benzoic or salicylic acids found in musts or wines although salicylic acid was widely used at the turn of the century. Sensi­ tive color tests are available for their detection (3, 5, 6, 22, 74). Just before W o r l d W a r II monochlor- and monobromacetic acids were used both in the United States and i n Europe, but they are now pro­ hibited. To detect organic bromide or chloride, l i q u i d - l i q u i d extraction is commonly used followed by destruction of the organic matter and use of the Volhard or other classical procedures. Ion-specific electrodes r

r

w

&

w

Downloaded by UNIV OF GUELPH LIBRARY on July 15, 2012 | http://pubs.acs.org Publication Date: November 1, 1974 | doi: 10.1021/ba-1974-0137.ch006

T

In Chemistry of Winemaking; Webb, A.; Advances in Chemistry; American Chemical Society: Washington, DC, 1974.

142

C H E M I S T R Y OF W I N E M A K I N G

can also be used as well as thin-layer chromatography (75). G L C , par­ ticularly of trimethylsilyl derivatives, can detect some additives—hydroxybenzoic acid, for example (60).

Downloaded by UNIV OF GUELPH LIBRARY on July 15, 2012 | http://pubs.acs.org Publication Date: November 1, 1974 | doi: 10.1021/ba-1974-0137.ch006

Methods Required in Winery

Operations

The procedures used in winery operations vary greatly, depending on the types of products produced and their market. A small winery producing only one type of red wine may need only a few different analyses. A winery producing grape juice, grape concentrate, table wines, dessert wines, special natural (flavored) wines, vermouth, fruit wines, high-proof spirits, and commercial brandy w i l l require many different types of analyses. Total A c i d . Simple titration procedures are used to determine total acidity. Problems arise because of the widely varying amounts of different acids in wines: tartaric, malic, citric, lactic, succinic, acetic, etc. Different pK„ values for these acids make it impossible to predetermine easily the correct p H of the endpoint. Since a strong base is being used to titrate relatively weak acids, the endpoint w i l l be greater than p H 7. In this country phenolphthalein (8.3) or cresol red (7.7) endpoints or a p H meter to 9.0 have been used (3, 6, 12, 76, 77); and the results are ex­ pressed as tartaric acid. The result at p H 7.7 X 1.05 approximately equals the result of titrating to p H 8.4. In Europe p H 7 is usually the endpoint, in France the results are expressed as sulfuric acid, and in Germany as tartaric or in milliequivalents (78). The possibility of determining the acids in wines from the titration curve using special equations has been extensively investigated in Portu­ gal by Pato and coworkers (79). T o keep the ionic force constant, ap­ propriate dilution is needed. Tartaric, malic, lactic, and succinic acid were determined in musts and wines. Fixed A c i d . The total acid (as tartaric) less the volatile acidity (as tartaric) is the fixed acidity. It is useful to make this calculation when one suspects activity of acid-reducing bacteria, as in the malo-lactic fermentation. Malic A c i d . This is seldom determined quantitatively in winery practice. However, qualitative paper chromatography is often done to follow malo-lactic fermentation. Using η-butyl alcohol and formic acid (80), the R values are: tartaric 0.28, citric 0.45, malic 0.51, ethyl acid tartrate 0.59, lactic acid 0.78, succinic 0.78, and ethyl acid malate 0.80. f

For quantitative results no completely satisfactory procedure is avail­ able. Enzymatic procedures (81) using L-malic dehydrogenase suffer from possible interference of tartaric acid (82). M a l i c acid can be fer-

In Chemistry of Winemaking; Webb, A.; Advances in Chemistry; American Chemical Society: Washington, DC, 1974.

Downloaded by UNIV OF GUELPH LIBRARY on July 15, 2012 | http://pubs.acs.org Publication Date: November 1, 1974 | doi: 10.1021/ba-1974-0137.ch006

6.

Wine Analysis

AMERINE

143

men ted by Schizosaccharomyces pombe (83). The colorimetric procedure with 2,7-naphthalene disulfonic acid (84) is perhaps the most practicable. T a r t a r i c A c i d . Quantitative measures of total tartrate are useful in determining the amount of acid reduction required for high acid musts and in predicting the tartrate stability of finished wines. Three procedures may be used. Precipitation as calcium racemate is accurate (85), but the cost and unavailability of L-tartaric acid are prohibitive. Precipitation of tartaric acid as potassium bitartrate is the oldest procedure but is some­ what empirical because of the appreciable solubility of potassium b i ­ tartrate. Nevertheless, it is still an official A O A C method ( 3 ) . T h e colorimetric metavanadate procedure is widely used (4, 6, 86, 87). Tan­ ner and Sandoz (88) reported good correlation between their bitartrate procedure and Rebelein's rapid colorimetric method (87). Potentiometric titration i n M e C O after ion exchange was specific for tartaric acid (89). 2

Lactic A c i d . Qualitative and even semiquantitative data are obtained by paper chromatography. Quantitative procedures where lactic acid is oxidized to acetaldehyde and the acetaldehyde determined colorimetrically are available (4,13, 22, 90). C i t r i c A c i d . This acid is rarely determined in wines since only a small amount is present. However, should it be used to modify the acid taste, limits would be imposed, and its determination would be necessary. Enzymatic (22, 91, 92) and colorimetric (93, 94) procedures are available. Simultaneous Determination of Acids. W i t h the advent of trimethylsilyl derivatives of the organic acids, G L C determination has been devel­ oped (60, 95) as a method for their detection and quantitative deter­ mination. p H . Because it affects the growth of microorganisms, color, taste, the ratio of free-to-total sulfur dioxide, and susceptibility to iron phos­ phate cloudiness, the p H is commonly measured as a guide to winery practice. Ordinary p H meters are used. Acetaldehyde. In routine winery operation acetaldehyde is seldom measured. However, i n the production of sherry, either by the film yeast or submerged culture processes, regular acetaldehyde determination is necessary. The principles of the classical Jaulmes and Espezel (3, 96, 97) pro­ cedure are still used. T o prevent copper-catalyzed oxidative changes, E D T A is added to remove copper (98). Isopropyl alcohol has the same effect (99). Colorimetric (78, 94, 100, 101, 102) and G L C (103, 104) procedures are becoming popular. H y d r o x y m e t h y l f u r f u r a l . W h e n must or wines containing fructose are heated, hydroxymethylfurfural is produced. It is easily detected qualitatively, and limits are placed on it i n Germany to prevent over­ heating of grape and fruit juices and i n Portugal to prevent artificial aging

In Chemistry of Winemaking; Webb, A.; Advances in Chemistry; American Chemical Society: Washington, DC, 1974.

Downloaded by UNIV OF GUELPH LIBRARY on July 15, 2012 | http://pubs.acs.org Publication Date: November 1, 1974 | doi: 10.1021/ba-1974-0137.ch006

144

C H E M I S T R Y OF W I N E M A K I N G

of port wine by heating. Paper and thin-layer chromatography and spectrophotometric procedures have been used (6, 23, 89, 131). Sorbic A c i d and Sorbates. In addition to the correction of the volatile acidity for sorbic acid already mentioned, both sorbic acid and sorbates must be determined directly. The colorimetric procedure of Jaulmes et al. (10) is appropriate: oxidation of sorbic acid to malonic dialdehyde and a red color developed by reaction with 2-thiobarbituric acid (4). Sorbitol and Mannitol. Sorbitol is present in fruits but not i n grapes. A method for its determination is required to detect illegal blending of fruit wines with grape wines. Mannitol is produced by bacterial spoilage. Sorbitol dehydrogenase and thin-layer chromatography have been used for their simultaneous determination ( 5 ) . Glycerol. Glycerol is seldom determined today except i n research work. G L C appears to be the method of choice (105) although the enzymatic procedure is direct and accurate (22, 106). Copper. In the presence of sulfur dioxide, copper-protein cloudiness may develop in white wines. Only small amounts of copper (about 0.3 to 0.5 mg/liter) cause cloudiness. Widespread use of stainless steel in modern wineries has reduced copper pickup, but many wineries routinely test their wines for copper. Atomic absorption spectrophotometry is the method of choice (51) although reducing sugars and ethanol interfere, and correction tables must be used (107). To reduce this interference, chelating and extracting with ketone is recommended (108). Lacking this equipment colorimetric procedures can be used, especially with d i ethyldithiocarbamate (3, 4, 6, 9,10, 22,109). Neutron activation analysis has been used for determining copper in musts (110). Iron. Excess iron in wines causes cloudiness, interferes with the color, and can impair flavor. T h e mechanism of ferric phosphate precipitation has been intensively studied, and numerous colorimetric methods have been developed. F o r routine purposes the color developed with thiocyanate is adequate (6,9), but many enologists prefer the orthophenanthroline procedures (3, 4, 6, 22). Meredith et al. (Ill) obtained essentially the same results for iron using 2,4,6-tripyridyl-s-triazine ( T P T Z ) to develop the color. Atomic absorption spectrophotometry can be used but, as with copper, corrections for reducing sugar and ethanol are necessary (51). Potassium. Quality standards for bottled wines now require a high degree of clarity. Even slight precipitates of potassium acid tartrate are considered detrimental. Whether wines are stabilized by cold treatment, long aging, or ion exchange, determination of their potassium content may be necessary. Precipitation as the acid tartrate (6) is widely used. H o w ­ ever, precipitation as potassium tetraphenylborate is used in Europe (4, 22). Flame photometry and atomic absorption spectrophotometry are

In Chemistry of Winemaking; Webb, A.; Advances in Chemistry; American Chemical Society: Washington, DC, 1974.

Downloaded by UNIV OF GUELPH LIBRARY on July 15, 2012 | http://pubs.acs.org Publication Date: November 1, 1974 | doi: 10.1021/ba-1974-0137.ch006

6.

AMERINE

Wine

Analysis

145

used by those with the necessary equipment (5, 22, 112). Sodium has been discussed i n connection with legal restrictions. Calcium. Excess calcium can occur in wines stored in concrete tanks or otherwise exposed to calcium (filter aids, calcium bentonite, etc.). After fortified wines are bottled, calcium tartrate may slowly precipitate. Complexometric titration with E D T A is the usual winery procedure for determining calcium (4, 6, 22, 113), but atomic absorption spectro­ photometry (51,53,112) and flame photometry and a rapid micro method based on oxalate precipitation (114) have been used successfully. Tannin. The phenolic compounds are of increasing importance i n winery operation to predict the proper fining agent and amount and method of fining as a measure of the age and flavor of the wine, as an indication of the degree of oxidation the wine w i l l tolerate, etc. The present standard methods (3, 4, 5, 6, 22) use the Folin-Ciocalteu reagent (115, 116). Gallic acid is used to prepare the standard curve at 765 nm. For recent methods for total phenolics and for separating the phenolic fractions, see Singleton and Esau (117) and P. Ribéreau-Gayon ( U S , 119). Histamine. The presence of small amounts of histamine in wines is now well established. These amounts are usually well below those that may have physiological effects. A method for determining histidine and histamine simultaneously is now available (120). The precision and sensitivity were 0.05 mg/liter and 0.025 /ug/ml and 0.7 mg/liter and 0.75 /J.g/ml, respectively. Color. Precise specification of color has become more important as many producers sell large volumes of specific wines to a national clientele. There is the additional problem of low color white wines (5). The usual procedure is to prepare an absorption curve in the visible spectrum. From this, the trichromatic coefficients are determined and the requisite three color parameters—luminence, dominant wavelength, and purity are obtained. T h e procedure is somewhat laborious, and shorter methods are often used. For white wines, changes i n absorption at 420 or 430 nm are adequate to detect browning changes. F o r red wines the absorbance at 420 and 520 nm is measured. The ratio of these corresponds roughly with changes i n tint or hue (dominant wavelength). The sum of the two absorbances is a rough measure of luminence (brightness). This method detects color changes i n most red wines (121) even though the human eye can detect even smaller differences (122). Methods Required in Research The research enologist requires many different procedures depending on the requirements of the experiment. Research on flavor constituents requires the most sophisticated techniques of modern organic chemistry.

In Chemistry of Winemaking; Webb, A.; Advances in Chemistry; American Chemical Society: Washington, DC, 1974.

Downloaded by UNIV OF GUELPH LIBRARY on July 15, 2012 | http://pubs.acs.org Publication Date: November 1, 1974 | doi: 10.1021/ba-1974-0137.ch006

146

C H E M I S T R Y OF W I N E M A K I N G

Among the compounds commonly determined i n research labora­ tories are diacetyl, 2,3-butandiol, glycerol, citramalic acid, amino acids (especially proline), histamine, ammonia, succinic acid, phosphate, ash, alkalinity of the ash, ethyl, acetate, methyl anthranilate, total volatile esters, higher alcohols (both total and individually) phenolic compounds, etc. A n elegant method for determining ethyl esters, capronate, caprylate, caprinate, and laurate using carbon disulfide extraction and G L C has been published (123). Quality standards for some flavor constituents w i l l eventually be developed—linalool for muscats, for example, and perhaps phenethanol for certain types of wine. Kahn and Conner (124) have published a rapid G L C method for phenethanol. It has been suggested (60) that detection of bacterial activity from the presence and amount of minor bacterial byproducts (arabitol, erythritol, and mannitol) may be useful. Based on G L C determination of carbonyls, esters, and higher alcohols, beers were accurately classified into three categories (125). Anthocyanin content has been determined quantitatively by using molar absorbance values for five anthocyanin pigments (126). Future

Trends

G L C , atomic absorption and mass spectrophotometry, enzymatic, and specific colorimetric procedures seem to be the likely candidates for routine use in the future. Automation w i l l certainly be common. G L C is now used to detect imitation muscat wines (127). Characteristic flavor byproducts of yeasts may be detected and measured. Multiple correla­ tion of the amounts of the more influential major and minor constituents with wine quality is the goal of such research. A simple apparatus for the simultaneous determination of the redox potential (two platinum elec­ trodes), p H , specific conductivity, oxygen, and carbon dioxide (ionspecific electrode) has been devised (128). Molecular oxygen i n wines has been determined by several procedures—polarography (129) and G L C being the latest. Enologists are now using a variety of analytical procedures. It is difficult to predict sources of new methods, but of those currently used the trend seems to be toward rapid, automated procedures, some based on colorimetry and others based on refractometry. G L C is already used widely, but, when coupled with special integrating printout systems, this method becomes even more attractive. H i g h pressure liquid chromatog­ raphy is useful for compounds not suitable for G L C . It is recommended for separating and accurately measuring phenolic acids and flavanoids in red wines (130). Thin-layer chromatography has also been used in research and, along with rapid scanning techniques, it may find wider

In Chemistry of Winemaking; Webb, A.; Advances in Chemistry; American Chemical Society: Washington, DC, 1974.

6.

AMERINE

Wine Analysis

147

Downloaded by UNIV OF GUELPH LIBRARY on July 15, 2012 | http://pubs.acs.org Publication Date: November 1, 1974 | doi: 10.1021/ba-1974-0137.ch006

application. Ion-specific electrodes also appear to be especially useful. Fluorometric analysis has been used for research but could find wider use if controls develop on certain constituents, such as histamine, where it is a method of choice. In the future new constituents may be analyzed for in an ecologically oriented society. Only a single paper on the presence of traces of urethane in diethylpyrocarbamate-treated wine was sufficient to ban its use. Under the Delaney amendment the zero-tolerance rule may require many new analytical procedures of very small amounts of certain components of wines derived directly or indirectly from additives. Obviously the future enologist w i l l have to be as sophisticated i n the chemical analytical laboratory as in sensory examination. Literature

Cited

1. Dujardin, J., Dujardin, L., Dujardin, R., "Notice sur les instruments de prècision appliqués à l'oenologie," Dujardin-Salleron, Paris, 1928. 2. Amerine, M. Α., J. Ass. Off. Agr. Chem. (1961) 44, 380. 3. "Official Methods of Analysis," 11th ed., Association of Official Agricultural Chemist, Washington, 1970. 4. "Recueil des Méthodes Internationales d'Analyse des Vins," Office Inter­ national de la Vigne et du Vin, Paris, 1962-1972. 5. Amerine, M. Α., Ough, C. S., "Wine and Must Analysis," Wiley, New York, 1974. (See also Encycl. Ind. Chem. Anal. 18, in press.) 6. Amerine, M. Α., "Laboratory Procedures for Enologists," Associated Stu­ dents Bookstore, Davis, 1970. 7. Amerine, Μ. Α., Wine Inst.Technol.Advis. Comm., June 14, 1971. 8. Amerine, Μ. Α., Berg, H. W., Cruess, W. V., "The Technology of Wine Making," 3rd ed., Avi, Westport, 1972. 9. Amerine, Μ. Α., Joslyn, Μ. Α., "Table Wines. The Technology of Their Production," University of California, Berkeley, Los Angeles, 1970. 10. Jaulmes, P., Mestres, R., Mandrou, B., Ann. Fals. Expert. Chim. (1964) 57, 119. 11. Joslyn, Μ. Α., "Methods in Food Analysis. Physical, Chemical and Instru­ mental Methods of Analyses," 2nd ed., Academic, New York, London, 1970. 12. "Uniform Methods of Analyses for Wines and Spirits," American Society of Enologists, Davis, 1972. 12a. Joslyn, Μ. Α., Amerine, Μ. Α., "Dessert, Appetizer and Related Flavored Wines. The Technology of Their Production," University of California, Berkeley, 1964. 13. Franck, R., Junge, C., "Weinanalytik. Untersuchung von Wein und ähnlichen älkoholischen Erzeugnissen sowie von Fruchtsäften," Carl Heymanns Verlag, Koln, 1970. 14. Hennig, K., Jakob, L., "Chemische Untersuchungsmethoden für Weinbereiter und Sussmosthersteller," 6th ed., Verlag Ulmer, Stuttgart, 1972. 15. Hess, D., Koppe, F., in "Handbuch der Lebensmittel Chemie," vol. VII, "Alkoholische Genussmittel," Springer-Verlag, Berlin, 1968. 16. Jaulmes, P., "Analyse des Vins," 2nd ed., Libr. Coulet, Dubois et Poulain, Montpellier, 1951. 17. Kourakou-Dragon, S., "Diethneis Methodoi Analyseos ton Glefkon ke Oinon," Ektyposis Institoutou Georgikis Mechanologias, Athens, 1971.

In Chemistry of Winemaking; Webb, A.; Advances in Chemistry; American Chemical Society: Washington, DC, 1974.

Downloaded by UNIV OF GUELPH LIBRARY on July 15, 2012 | http://pubs.acs.org Publication Date: November 1, 1974 | doi: 10.1021/ba-1974-0137.ch006

148

CHEMISTRY OF WINEMAKING

18. Mori, L., "Metodi Ragionali di Analisi nella Moderna Technica Enologica," 2nd ed., Luigi Scialpi Editore, Rome, 1967. 19. Nilov, V. I., Skurikhin, I. M., "Khimiya Vinodeliya i Kon'yachnogo Proizvodstva," 2nd ed., Pishchepromizdat, Moscow, 1967. 20. Rankine, B. C., "Principles and Pitfalls of Winery Analyses," Gawler Adult Education Centre, Oenology Course for Winemakers, 1961. 21. Rentschler,H.,Tanner, H., "Anleitung für die Getränke-Analyse," 6th ed., Eidg. Forschungsanstalt, Wädenswil, 1971. 22. Ribéreau-Gayon, J., Peynaud, E., Sudraud P., Ribéreau-Gayon, P., "Analyse et Contrôle des Vins," Dunod, Paris, 1972. 23. Vogt, E., Bieber, H., "Weinchemie und Weinalyse," 3rd ed., Verlag E. Ulmer, Stuttgart, 1970. 23a. Jakob, L., All. Deut. Weinfachztg (1971) 107, 1163. 24. Rebelein, H., Allg. Deut. Weinfachztg. (1971) 107, 590. 25. Schmitt, Α., Allgem. Deut. Weinfachztg. (1971) 107, 1962. 26. Schmitt, Α., Deut. Weinbau. (1972) 27, 57. 27. Sarris, J., Morfaux, J. N., Dupuy, P., Hertzog, D., Ind. Aliment. Agr. (1969) 86, 1241. 28. Wanger, O., Mitt. Geb. Lebensmunters. Hyg. (1969) 60, 271. 29. Lie, S., Haukeli, A. D., Gether, J. J., Brygmesteren. (1970) 27, 281. 30. Tanner, H., Brunner, E. M., Mitt. Geb. Lebensmittelunters. Hyg. (1964) 55, 480. 31. Dyer, R. H., J. Ass. Off. Anal. Chem. (1972) 55, 564. 32. Martin, G. E., Caggiano, G., Beck, J. E., J. Ass. Off. Agr. Chem. (1963) 46, 297. 33. Pilone, G. J., Rankine, B. C., Hatcher, C. J., Aust. J. Wine, Brew. Spirit Rev. (1972) 91, 62 34. Jakob, L., Rebe Wein. (1971) 25, 44. 35. Postel, W., Drawert, F., Maccagnan, G., Chem., Mikrobiol., Technol. Lebensm. (1971) 1, 11. 36. Rebelein, H., Steinert, H., Ger. Offen. (1971) 2, 126. 37. Wucherpfennig, K., Bretthauer, G., Wein-Wiss. (1971) 26, 405. 38. Sarris, J., Morfaux, J. N., Dervin, L., Connais. Vigne Vin. (1970) 4, 431. 39. Wucherpfennig, Κ., Jahresber. Hess. Forschungsanstalt Wein. Obst. Gartenbau. Geisenheim. (1972) 1971, 46. 40. Deibner, L., Bernard, P., Chim. Anal. Paris (1970) 54, 412. 41. Jakob, L., Weinblatt. (1970) 64, 461. 42. Lay, Α., Mitt. Rebe Wein, Obst. Fruchteverwert. (1970) 20, 85. 43. Rebelein, H., Mitt.-Bl. GDCH-Fachgr. Lebensmittelchem. gerichtl. Chem. (1969) 23, 107. 44. Tanner, H., Sandoz, M., Schweiz. Z. Obst. Weinbau. (1972) 108, 251. 45. Stanescu, C., Bull. O.I.V. (Off. Intern. Vigne Vin) (1972) 45, 785. 46. Junge, C., Deut. Lebensm. Rundsch. (1970) 66, 421. 47. Lotti, G., Baldacci, P. V., Riv. Viticolt,Enol.(1970) 23, 262. 48. Rebelein, H., Bull. O.I.V. (Off. Int. Vigne Vin) (1968) 41, 344. 49. Rebelein, H., Deut. Lebensm. Rundsch. (1967) 63, 233. 50. Mack, D., Berg, H., Deut. Lebensm. Rundsch. (1972) 68, 262. 51. Varjú, M., Z. Lebensm. Unters. Forsch. (1972) 148, 268. 52. Brunn, S., Bonnemaire, J. P., Rev. Franç. Oenol. (1971) 43 (3), 12. 53. Polo, M. C., Garrido, M. D., Llaguno, C., Garrido, J., Rev. Agroquím. Technol. Aliment. (1969) 9, 600. 54. Raik, S. Ya., Kryzhanovskaya, E. Kh., Sadovod. Vinograd. Vinodel. Mold. (1970) 25 (3), 36. 55. Barros, Μ. Η. Β.,ν. de, Estud., Notas Relatorios. Porto (1971) 7, 59. 56. Hadorn, H., Zürcher, K., Ragnarson, V., Mitt. Geb. Lebensmittelunters. Hyg. (1967) 58, 1.

In Chemistry of Winemaking; Webb, A.; Advances in Chemistry; American Chemical Society: Washington, DC, 1974.

Downloaded by UNIV OF GUELPH LIBRARY on July 15, 2012 | http://pubs.acs.org Publication Date: November 1, 1974 | doi: 10.1021/ba-1974-0137.ch006

6.

AMERINE

Wine Analysis

149

57. Hrazdina, G., J. Agr. Food Chem. (1970) 18, 243. 58. Schmidt-Hebbel, H., Michelson, W., Masson, L., Steltzer,H.,Z. Lebensm. Unters. Forsch. (1968) 137, 169. 59. Drawert, F., Kupfer, G., Fresenius Z. anal. Chem. (1965) 211, 89. 60. Ribéreau-Gayon, P., Bertrand, Α., Vitis. (1972) 10, 318. 61. Ough, C. S., Cooke, G. M., Wines Vines. (1966) 27 (8), 27. 62. Parenthoen, Α., Ann. Fals. Expert. Chim. (1972) 65, 279. 63. Schopfer, J. F., Regamey, R., Rev. SuisseViticult.Arboricult. (1971) 3, 107. 64. Rotolo, Α., Riv. Viticolt.Enol.(1972) 25, 301. 65. Woidich, H., Gnauer, H., Tunka, J., Z. Lebensm. Unters. Forsch. (1971) 147, 284. 66. Matthey, E., Rentschler, H., Schopfer, J. F., Mitt. Geb. Lebensmittel­ unters. Hyg. (1971) 62, 101. 67. Lemperle, E., Kerner, E., Fresenius Z. anal. Chem. (1969) 247, 49. 68. Vitali, G., Ind. Aliment. Pinerolo, Italy (1970) 9, 71. 69. Bandion, F., Mitt. Rebe Wein, Obstbau Fruchteverwert. Klosterneuburg (1969) 19, 37. 70. Hara, S., Murakami, H., Marukāwa, K., Omata, Y., Sakakura, I.,J.Ferment.Technol.(1970) 48, 616. 71. Wunderlich, H., J. Ass. Off. Anal. Chem. (1972) 55, 557. 72. Hoppe, H., Romminger, K., Nahrung (1969) 13, 227. 73. Bates, B. L., J. Ass. Off. Anal. Chem. (1970) 53, 775. 74. English, E., Analyst (1959) 84, 465. 75. Haller, H. E., Junge, J., Deut. Lebensm. Rundsch. (1971) 67, 231. 76. Guymon, J. F., Ough, C. S., Amer.J.Enol.Viticult.(1962) 13, 40. 77. Wong, G., Caputi, Jr., Α., Amer. J. Enol. Viticult. (1966) 17, 174. 78. Owades, J. L., Dono, J. M., J. Ass. Off. Anal. Chem. (1968) 51, 148. 79. Pato, M. A. da S., Pato, M. H. M. L. da S., De Vinea et Vino Port. Doc. (1972) 6 (3), 1. 80. Kunkee, R. E., Wines Vines (1968) 49 (3), 23. 81. Mayer, K., Busch, I., Mitt. Geb. Lebensmittelunters. Hyg. (1963) 54, 60. 82. Poux, G., Ann.Technol.Agr. (1969) 18, 359. 83. Peynaud, E., Lafon-Lafourcade, S., Ann. Technol. Agr. (1965) 14, 49. 84. Tarantola, C., Castino, M., Ann. Fac. Sci. Agr. Univ. Torino. (1962) 1, 137. 85. Martiniere, P., Sudraud, P., Connais. Vigne Vin. (1968) 2, 41. 86. Hill,G.,Caputi, Jr., Α., Amer. J.Enol.Viticult.(1970) 21, 153. 87. Rebelein, H., Mitt.-Bl. GDCH-Fachgr. Lebensmittelchem. gerichtl. Chem. (1970) 24, 14. 88. Tanner, H., Sandoz, M., Schweiz, Z. Obst. Weinbau. (1972) 108, 251. 89. Fal'kovich, Y. E., Gun'ko, G. P., Avenes'yants, R. V., Vinodel. Vinograd. SSSR (1972) 32 (4), 38. 90. Pilone, G. J., Kunkee, R. E., Amer. J. Enol. Viticult. (1970) 21, 12. 91. Mayer, K., Pause, G., Mitt. Geb. Lebensmittelunters. Hyg. (1965) 56, 454. 92. Mayer, K., Pause, G., Lebensm.-Wiss.Technol.(1969) 2, 143. 93. Addeo, F., Sci.Technol.Aliment. (1972) 2, 87. 94. Rebelein, H., Deut. Lebensm. Rundsch. (1967) 63, 337. 95. Martin, G. E., Sullo, J.G.,Schoeneman, R. L., J. Agr. Food Chem. (1971) 19, 995. 96. Guymon, J. F., Wright, D. L., J. Ass. Off. Anal. Chem. (1967) 50, 305. 97. Jaulmes, P., Espezel, P., Ann. Fals. Fraudes. (1935) 28, 325. 98. Jaulmes, P., Hamelle, G., Ann. Fals. Expert. Chim. (1961) 54, 338. 99. Burroughs, L. F., Sparks, A. H., Analyst (London) (1961) 86, 381. 100. Heintze, K., Braun, F., Z. Lebensm. Unters. Forsch. (1970) 142, 40. 101. Rebelein, H., Deut. Lebensm. Rundsch. (1970) 66, 6.

In Chemistry of Winemaking; Webb, A.; Advances in Chemistry; American Chemical Society: Washington, DC, 1974.

150 102. 103. 104. 105. 106. 107. 108. 109. 110. Downloaded by UNIV OF GUELPH LIBRARY on July 15, 2012 | http://pubs.acs.org Publication Date: November 1, 1974 | doi: 10.1021/ba-1974-0137.ch006

111. 112. 113. 114. 115. 116. 117. 118. 119. 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 131.

CHEMISTRY OF WINEMAKING

Then, R., Radler, F., Ζ. Lebensm. Unters. Forsch. (1968) 138, 163. Andre, L., Ann.Technol.Agr. (1966) 15, 159. Morrison, R. L., Amer. J. Enol. Viticult. (1962) 13, 159. Ough, C. S., Fong, D., Amerine, Μ. Α., Amer. J. Enol. Viticult. (1972) 23, 1. Mayer, K., Busch, I., Mitt. Geb. Lebensmittelunters. Hyg. (1963) 54, 297. Caputi, Jr., Α., Ueda, M., Amer. J. Enol. Viticult. (1967) 18, 66. Cameron, A. G., Hackett, D. R., J. Sci. Food Agr. (1970) 21, 535. Strunk, D.H.,Andreasen, Α. Α., J. Ass. Off. Anal. Chem. (1967) 50, 334, 338. Baraldi, D., Nikolai, K., Stehlik, G., Altmann, H., Kaindl, K., Inst. Biol. Agr. Seibersdorf Reactor Centre (1968) Spr. No. 17. Meredith, M. K., Baldwin, S., Andreasen, Α. Α., J. Ass. Off. Anal. Chem. (1970) 53, 12. Bergner, K. G., Lang, B., Deut. Lebensm. Rundsch. (1971) 67, 121. Blouin, J., Llorca, L., Leon, P., Connais. Vigne Vin. (1971) 5, 99. Weger, Β., Riv. Viticolt.Enol.(1968) 21, 441. Seider, A. I., Datunashvili, Ε. N., Vinodel. Vinograd. SSSR (1972) 32 (6), 31. Singleton, V. L., Rossi, Jr., J. Α., Amer. J.Enol.Viticult.(1965) 16, 144. Singleton, V. L., Esau, P., "Phenolic Substances in Grapes and Wines, and Their Significance," Academic, New York, London, 1969. Ribéreau-Gayon, P., Rev. Franç. Oenol. (1971) 11, 25. Ribéreau-Gayon, P., Chim. Anal. (1970) 52, 627. Plumas, B., Sautier, C., Ann.Fals.Expert. Chim. (1972) 65, 322. Ough, C. S., Berg, H. W., Chichester, C. O., Amer. J. Enol. Viticult. (1962) 13, 32. Bere, H. W., Ough, C. S., Chichester, C. O., J. Food Sci. (1964) 29, 661. Koch, J., Hess, D., Gruss, R., Z. Lebensm. Unters. Fosch. (1971) 147, 207. Kahn, J.H.,Conner, Η. Α., J. Ass. Off. Anal. Chem. (1972) 55, 1155. Postel, W., Drawert, F., Adam, L., Chem., Mikrobiol.,Technol.Lebensm. (1972) 1, 169. Niketić-Aleksić, G. K., Hrazdina, G., Lebensm. Wissen.Technol.(1972) 5, 163. Spanyár, P., Kavei, E., Blazovich, M., Fr. Viticole (Montpellier) (1972) 4, 131. Litchev, V., Goranov, N., Albert, H., Yanakiev, M., Bull. O.I.V. (Off. Intern. Vigne Vin.) (1972) 45, 1059. Tsuyboul'kova, L. P., Balanoutse, A. P., Khramov, Α. V., Nilov, Β. I., Datounaschvili, Ε. N., Vinodel. Vinograd. SSSR (1972) 32 (4), 30. Charalambous, G., Bruckner, K. J., Hardwick, W. Α., Linnebach, Α., Master Brew. Ass. Amer. Tech. Quart. (1973) 10, 74. Izard-Verchére,C.,Viel, C., Bull. Soc. Chim. Fr. (1972) 5, 2089.

RECEIVED May 29,

1973.

In Chemistry of Winemaking; Webb, A.; Advances in Chemistry; American Chemical Society: Washington, DC, 1974.